[louiskottmann]

The linked article does not speak of tests, it speaks of a team that failed to properly review an LLM refactor then proceeds to blame the tooling.

LLMs are good at writing tests in my experience.